Hi Upayavira,
It sounds like an error I also got. It seems that Lucene doesn't release
some Objects and keeps increasing memory consumption. And the end a parsing
error occurs.
My solution was to raise the memory heap size to the JVM with the
CATALINA_OPTS variable : "CATALINA_OPTS=-Xmx128m" (set 128 Mo of heap size,
the default value is 64 Mo).
Then to raise the default heap size in cocoon.xconf :
<store-janitor class="org.apache.cocoon.components.store.StoreJanitorImpl"
logger="core.store.janitor">
...
<!-- Indicates the limit of the jvm memory consumption. The default max
heapsize for Sun's JVM is 64Mb -->
<parameter name="heapsize" value="128000000"/>
<---------
...
</store-janitor>
And it worked...
This turn-around worked also :
<!-- Memory Storing: -->
<cache-transient class="org.apache.cocoon.components.store.MRUMemoryStore"
logger="core.store.transient">
<!-- Indicates how many objects will be hold in the cache.
When the number of maxobjects has been reached. The last object in the
cache will be thrown out.
<parameter name="maxobjects" value="100"/>
-->
<parameter name="maxobjects" value="50"/>
...
</cache-transient>
Or an other value of maxobjects. But there is a real decrease of
performance...
But is's just a "turn-around"... Not a solution.
Olivier
----- Original Message -----
From: "Upayavira" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Tuesday, March 18, 2003 12:47 AM
Subject: Lucene index building
> I have built a site which I want to index with Lucene.
>
> I am using the create-index.xsp file in the $COCOON-ROOT/search directory
to
> build my index.
>
> I have added the following to cocoon.xconf:
>
> <cocoon-crawler logger="core.search.crawler">
> <exclude>.*/search/.*</exclude>
> <link-view-query>cocoon-view=lucene-links</link-view-query>
> </cocoon-crawler>
>
> <lucene-xml-indexer logger="core.search.lucene">
> <store-fields>body</store-fields>
> <content-view-query>cocoon-view=lucene-content</content-view-query>
> </lucene-xml-indexer>
>
> I've set up a view lucene-links which works, giving back just links from a
page.
> I've set up a view lucene-content just giving back the content. The
content is like:
> <page>
> <links>....list of links</links>
> <body>... the body content ...</body>
> </page>
>
> I have had it partially working (indexing both links and body), but now
whenever I
> run create-index, it fails with a Cannot parse!:
org.xml.sax.SAXParseException:
> Premature end of file.
>
> Any ideas what I might be doing wrong?
>
> Regards, Upayavira
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
